https://github.com/oleksiygarnik/vehicles-trajectory-detection

A comprehensive solution for detecting and tracking vehicle trajectories using YOLO, Kalman Filter, and the Hungarian Algorithm. Includes a Flask web application for easy video upload and processing.
https://github.com/oleksiygarnik/vehicles-trajectory-detection

computer-vision flask hungarian-algorithm kalman-filter opencv python trajectory-tracking vehicle-detection video-processing yolo

Last synced: 5 months ago
JSON representation

A comprehensive solution for detecting and tracking vehicle trajectories using YOLO, Kalman Filter, and the Hungarian Algorithm. Includes a Flask web application for easy video upload and processing.

• Host: GitHub
• URL: https://github.com/oleksiygarnik/vehicles-trajectory-detection
• Owner: oleksiygarnik
• License: mit
• Created: 2020-07-29T14:30:26.000Z (almost 5 years ago)
• Default Branch: master
• Last Pushed: 2024-05-30T13:46:07.000Z (about 1 year ago)
• Last Synced: 2024-11-12T09:13:26.866Z (7 months ago)
• Topics: computer-vision, flask, hungarian-algorithm, kalman-filter, opencv, python, trajectory-tracking, vehicle-detection, video-processing, yolo
• Language: Python
• Homepage:
• Size: 12.7 KB
• Stars: 4
• Watchers: 2
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Vehicles Trajectory Detection

This repository provides a comprehensive solution for detecting and tracking vehicle trajectories in video footage. The system utilizes advanced computer vision techniques including YOLO for object detection, Kalman Filter for prediction, and the Hungarian Algorithm for tracking. Additionally, it includes a Flask web application for uploading and processing videos, with results visualized and stored for further analysis.

## Features

- **YOLO Object Detection**: Utilizes the YOLO (You Only Look Once) algorithm for detecting vehicles in each frame of the video.

- **Kalman Filter**: Predicts the future position of detected vehicles, improving the accuracy of tracking.

- **Hungarian Algorithm**: Assigns detected objects to existing tracks, ensuring consistent tracking across frames.

- **Flask Web Application**: Provides a user-friendly interface for uploading videos and viewing results.

- **Event-Driven Architecture**: Uses events to manage video processing tasks and communicate results.